Bug 445423

Summary: hdf5 long double tests fail on ppc64
Product: [Fedora] Fedora Reporter: Orion Poplawski <orion>
Component: hdf5Assignee: Orion Poplawski <orion>
Status: CLOSED CURRENTRELEASE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: low Docs Contact:
Priority: low    
Version: rawhideCC: pertusus
Target Milestone: ---Keywords: Tracking
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: hdf5-1.8.7-2 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-11-22 22:20:41 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 238953    

Description Orion Poplawski 2008-05-06 19:55:00 UTC 
See http://koji.fedoraproject.org/koji/getfile?taskID=596775&name=build.log

Can remove some of the failures by changing CFLAGS to "-m64", but not all:

Testing soft normalized long double -> long conversions               *FAILED*
Testing soft normalized long double -> unsigned long conversions      *FAILED*
Testing soft long -> long double conversions                          *FAILED*
Testing soft unsigned long -> long double conversions                 *FAILED*
Comment 1 Orion Poplawski 2008-05-06 20:16:34 UTC 
-fno-strict-aliasing doesn't help.
Comment 2 Kevin Kofler 2008-05-09 15:51:06 UTC 
Why don't you just %ifnarch ppc64 the call to make check?
Comment 3 Orion Poplawski 2008-05-09 16:12:40 UTC 
Well, I rather not ship a broken library.  I'm still in communication with
upstream.  Looks like long double support may be considered experimental at this
point, so I may disable the check on ppc64 for now.
Comment 4 Orion Poplawski 2008-11-16 02:55:53 UTC 
Shipping hdf5 on pp64 with tests disabled.  Keep testing new releases to see if this gets fixed.